Persuasion

The field experiments on whether campaign contact changes a vote, and where the effect has shown up.

Oct 17, 20262 min read

ne of the things a political campaign tries to do is persuade voters to change their vote. What are the conditions under which that actually happens?

Kalla & Broockman (2018)

A meta-analysis of 40 field experiments, plus nine new canvassing experiments the authors ran themselves, on whether campaign contact changes a general-election vote.

Unsupported block: pub.terrain.quote

The effects that do appear turn up where the ballot carries no party cue: primaries, ballot measures.

The near-election contact studies average below zero, so the published record may still be the optimistic end.

Coppock, Hill & Vavreck (2020)

Forty-nine real 2016 US presidential advertisements, tested as 59 separate randomised experiments on 34,000 people.

Unsupported block: pub.terrain.quote

Synthesis

Every number here is from the United States, in a general election, in the last few weeks before the vote.

This is a finding about the campaign’s last weeks. It is not a finding about whether people change their minds — the same literature has voters switching between elections in large numbers.

No comparable field-experiment record exists for Australia or for Victoria.

Nothing here measures the effect of a conversation between people who know each other, which is a different intervention from a knock by a stranger.

It is not known how much of the measured zero is about the message and how much is about the fact that by the last weeks most voters have already decided.

Linked notes

/notes/turnout-tactics

/notes/why-voters-defect

/notes/what-counts-as-evidence

Did you enjoy this article?

Recommend it — Standard Reader surfaces well-loved writing to more readers across the network.

Terrain @terrain.pub